Return to BSD News archive
Path: sserve!manuel.anu.edu.au!munnari.oz.au!constellation!convex!convex!cs.utexas.edu!swrinde!gatech!udel!sbcs.sunysb.edu!stark.UUCP!gene From: gene@stark.uucp (Gene Stark) Newsgroups: comp.os.386bsd.questions Subject: Re: Problems with printer Date: 3 Mar 93 19:16:25 Organization: Gene Stark's home system Lines: 45 Message-ID: <GENE.93Mar3191625@stark.stark.uucp> References: <1993Mar3.110942.2528@luthien.in-berlin.de> NNTP-Posting-Host: stark.uucp In-reply-to: wagner@luthien.in-berlin.de's message of Wed, 3 Mar 93 11:09:42 GMT >I've got a Deskjet 500 and am using the interruptless line printer driver >which works quite well with pure text files. However, since I have got >to convert all my TeX jobs into graphics there are quite often really >big files and it's really no fun watching them printed out line >by line so slowly that you won't believe it. Is there anybody out >there who has managed to achieve reasonable performance even with >printing binary data? OK, I have been seeing a lot of stuff flying by about the printer driver I must say that I don't really understand it, but I'd like to. I am using a Diconix 150 inkjet printer that I originally bought for my laptop. I began by using the stock lpt driver with Eric Haug's patches. I got stuff to print about half the time, but the probe routines failed to recognize the printer when it was off on bootup and sometimes even when it was on. Finally, about a month ago I got fed up. I tried some of the interruptless drivers posted here, but they didn't work for me out of the box, so I didn't try to debug them. Instead, I took the modified stock driver, deleted all code whose purpose I didn't understand, and fixed what looked to me like various race conditions. Now I have a driver that always manages to attach the printer on bootup, and doesn't have the "still open" problem I had with the previous one. I'm not saying it's clean, I'm just saying it works for me. I regularly print TeX DVI files converted to 96dpi Epson-compatible graphics, and I have never had any problem due to this. When I see people complain about binary files I don't really know what to make of this. Can others elaborate? Maybe we can pool our experience and come up with a printer driver that really works. As an aside, I have now also taken a look at the fd driver and the wd driver, and I have to say that it is beyond me how they were ever debugged to their current level of stability. The code appears very convoluted, I can't really convince myself that they are not full of race conditions, and my assessment is that they are really badly in need of a complete rewrite. The only thing is, I have only one disk and one system, and I am using it for other work. I am not going to debug a wd driver on my data. Also, many of the problems seem to have to do with disparity in hardware. Whoever takes on a rewrite will need to have access to various pieces of hardware, or a tightly knit group of testers. - Gene Stark -- stark@cs.sunysb.edu